1. Field of the Invention
The present invention relates to animating image editing apparatus and animating image editing method it for editing a coded animating image.
2. Related Background Art
Recently, coding methods of coding a digital animating image at a high compression ratio by using an interframe correlation have been standardized as international methods. There is an MPEG as a representative method. According to the MPEG, an interframe correlation is maximally used, a motion compensation is performed with reference to frames before and after a frame to be coded, and after that, an interframe differential coding is executed. FIG. 5 shows an example of the MPEG coding. In the coding frame types of the MPEG, I frame, P frame, and B frame have been coded as shown in the diagram. The I frame is the independent frame in an intraframe code and one frame is inserted per 15 frames. The P frame is a forward motion predictive frame and is an interframe code for performing the motion compensation by referring to the one-preceding P frame or the I frame. The B frame is an interframe predictive code for performing the motion compensation by predicting from both directions of the P and I frames before and after the B frame. The motion compensation is performed on a unit basis of a block which is called a macro block (MB) as shown in FIG. 6. The MB is a unit such that an animating image is digitized at a sampling ratio of 4:1:1 of Y (luminance component) and Cb and Cr (color difference components) and, after that, Y (16×16 pixels), Cb (8×8 pixels), and Cr (8×8 pixels) are set to a block. The motion compensation is executed in the B and P frames. Since the blocks having a strong correlation have to be searched (vector search) on a macro block unit basis, there is a drawback such that the processes of a coding unit are complicated as compared with those of a decoding unit. Particularly, when the coding is executed only by software, it is difficult to execute it in a real-time manner. It is, therefore, a present situation that only the decoding is performed by software. When the coding unit is realized by hardware, there is a draw back such that the circuit scale is extremely large.
As another problem, there is a problem at the time of the edition. Fundamentally, since the interframe difference is coded, it is difficult to perform a cut edition between arbitrary frames. For example, when an animating image is divided into two sequences at the position of {circle around (1)} in FIG. 5, there is a drawback such that the last frame cannot be decoded in the preceding sequence and the frames up to the next I frame cannot be decoded in the later sequence.
As a first method of solving the foregoing problems, there is a method of decoding all of the coded frame images and editing the decoded frame images after that.
The above method, however, has a drawback such that a memory of a large capacity is necessary to once accumulate the images after they were decoded.
In the above method, there is also a problem such that since all of the frame images need to be decoded irrespective of a fact such that the frame to be edited is a part of animating image and the decoded images are again encoded to accumulate them after that, the image deterioration is large.
As a second method of solving the foregoing problems, there is also an animating image processing apparatus using the intraframe coding method. For instance, there is a motion JPEG in which the JPEG system which is a coding of a color still image is independently applied to each frame of an animating image. In case of the motion JPEG, since the coding is also the intraframe coding, it is not so complicated as in the MPEG and a frame edition can be also freely performed. However, since the interframe correlation is not used, there is a drawback such that a compression efficiency is remarkably worse than that of the interframe coding.